A landmark therapeutic breakthrough was the preemptive use of antiviral drugs including ganciclovir (GCV) and foscarnet (FOS) to control cytomegalovirus (CMV) viremia in hematopoietic cell transplant (HCT) recipients. These drugs have remarkable properties to limit progression from viremia to disease but also have significant side effects, including exacerbating neutropenia, delay of immune reconstitution, and renal toxicity in the case of FOS. To overcome using these antiviral agents, we proposed a peptide vaccine to stimulate the host immune response to CMV. In the last funding period we conducted a Phase 1 trial of the peptide vaccine with or without Pfizer PF03512676 adjuvant that confirmed the safety of 2 different formulations, with evidence of immunologic activity based upon flow cytometry analysis of CMV-pp65- specific T cells. A limitation of the peptide vaccine is its restriction to individuals with the HLA-A2 haplotype which narrows the target population. While we are seeking to confirm a therapeutic benefit of the peptide vaccine in HCT recipients, broadening the population that would benefit from a vaccination strategy is best approached by employing a delivery vector that expresses full length CMV antigens. We have used modified vaccinia Ankara (MVA) virus for that purpose in collaboration with NCI-NExT to manufacture a candidate for clinical testing. In this revised application, we propose 3 clinical trials that will comprehensively evaluate our clinical vaccine candidate. In Aim 1, a safety study (Trial 1) will be conducted in healthy volunteers stratified by age and poxvirus exposure who are equivalent health-wise to immunocompetent HCT donors as required by the FDA and will establish the safety and dose level for a 2nd trial in HCT. In Aim 2, a double-blind Phase 2 trial is described in which 150 HLA matched related HCT donors (MRD) will be randomized equally to receive either vaccine or placebo. Immune responses and clinical correlates of vaccine-stimulated protection will be followed in recipients whose donors are enrolled in Trial 2. In Aim 3 (Trial 3), we propose to directly immunize 150 HCT recipients undergoing unrelated (8/8 HLA matches) donor (URD) HCT after gaining safety experience by following the clinical course of 30 recipients in Trial 2. This approach is justified since these recipients are at high-risk for CMV reactivation and do not have consistently accessible donors. Trials 2 and 3 will each enroll 150 HCT subjects, randomized equally to vaccine or placebo and both will have 90% power to detect a 50% reduction in viremia (60% to 30%) in the vaccine arm. Safety monitoring employing defined stopping rules will guard against adverse consequences for recipients in both trials. Combined results from carrying out both Trials 2 and 3 will provide a precise immunoprophylaxis strategy for control of CMV disease in the HCT setting. We will discover the optimal formula for immunizing both donors and recipients to prevent CMV viremia and minimize usage of potentially toxic antiviral agents. Success with this CMV vaccine would have a significant clinical and cost-saving effect on recipient management after MRD or URD HCT.